INDEX

Explanations

phrases indicating functionality or usage of objects or devices

oai_token-act-pair · gpt-4o-mini Triggered by @bot

be used for

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

BagLayout

-0.54

 getRule

-0.44

 equilíbrio

-0.44

 betweenstory

-0.44

PerformLayout

-0.44

modelBuilder

-0.44

[++

-0.44

 kasarigan

-0.42

 setDate

-0.42

LocationManager

-0.41

POSITIVE LOGITS

GEBURTSDATUM

0.43

 wireType

0.42

MessageTagHelper

0.40

Either

0.40

 disambiguazione

0.40

OGND

0.39

 intercamb

0.39

 húmedo

0.39

 رز

0.37

awsze

0.37

Activations Density 0.002%

phrases indicating functionality or usage of objects or devices

be used for

No Comments

No Known Activations